Authors

Nemanja Rakicevic and Petar Kormushev

Robot Intelligence Lab, Imperial College London

Abstract

Learning complex physical tasks via trial-and-error is still challenging for high-degree-of-freedom robots. Greatest challenges are devising a suitable objective function that defines the task, and the high sample complexity of learning the task. The greatest challenge is optimally selecting the next trials to evaluate while improving sample efficiency. We propose a novel active learning framework, consisting of decoupled task model and exploration components, which does not require an objective function. The task model is specific to a task and maps the parameter space, defining a trial, to the trial outcome space. The exploration component enables efficient search in the trial-parameter space to generate the subsequent most informative trials, by simultaneously exploiting all the information gained from previous trials and reducing the task model's overall uncertainty. We analyse the performance of our framework in a simulation environment and further validate it on a challenging bimanual-robot puck-passing task. Results show that the robot successfully acquires the necessary skills after only 100 trials without any prior information about the task or target positions. Decoupling the framework's components also enables efficient skill transfer to new environments which is validated experimentally.

Full paper

Source code

Experiment video

Experiment Results and Visualisation